Evaluating the mid-infrared bi-spectral index for improved assessment of low-severity fire effects in a conifer forest

Remote sensing products provide a vital understanding of wildfire effects across a landscape, but detection and delineation of low- and mixed-severity fire remains difficult. While data provided by the Monitoring Trends in Burn Severity project (MTBS) are frequently used to assess severity in the United States, alternative indices can offer improvement to the measurement of low-severity fire effects and should be considered for future product development and adoption. This research note evaluated one such alternative, the mid-infrared bi-spectral index (MIRBI), which was developed in savannah ecosystems to isolate spectral changes caused by burning and reduce noise from other factors. MIRBI, differenced MIRBI (dMIRBI) and burn severity indices used by MTBS were assessed for spectral optimality at distinguishing severity and the ability to separate unburned and burned canopy in a conifer forest. The MIRBI indices were better at isolating changes caused by burning and demonstrated higher spectral separability, particularly at low severity. These findings suggest that MIRBI indices can provide an enhanced alternative or compliment to current MTBS products in high canopy cover forests for applications such as discernment of fire perimeters and unburned islands, as well as identification of low-severity fire effects.